01 Executive Summary — Acts (English → Mandarin)

Curriculum: Acts 1–28 | Core passage: Acts 2:1–41 (Pentecost) | Destination: Mandarin Chinese (Simplified), CUV 神版 register

Why it matters

Acts is the narrative hinge between the Gospels and the epistles, and it stages — in dramatized, story form rather than didactic argument — nearly every doctrine that the destination culture’s dominant religious substrate (Chinese folk religion, Buddhism, Daoism, Confucianism) offers a superficially similar but theologically incompatible alternative to. Where Romans states doctrine, Acts narrates it: a spirit fills people (Pentecost), a sorcerer tries to buy spiritual power (Simon Magus), a crowd deifies human agents of a miracle (Lystra, Malta), and a movement is condemned to a name — “the Way” — that collides head-on with Daoist vocabulary. Getting these translation decisions right is not a stylistic nicety; it determines whether Mandarin readers encounter the text’s actual claims or unconsciously reassimilate them into mediumship, karma-merit, or divination frameworks already present in their cultural formation.

Key findings

• 101 enforced glossary terms are now recorded in assets/translation_memory.json: 40 baseline Romans terms reused exactly (God, Jesus, Lord, Holy Spirit, Grace, Righteousness, Justification, Salvation, Sanctification, Resurrection, and 30 others), plus 61 new terms Acts introduces across its full 28 chapters.
• 19 Critical-risk and 30 High-risk terms (49 of 101 total, ~49%) require mandatory human theologian review on every occurrence — a higher proportion than the Romans baseline, driven by Acts’ concentration of narrative scenes (Pentecost, Simon Magus, Stephen’s vision, Paul’s conversion, Lystra, Jerusalem Council, Philippian jailer, Areopagus, Malta) that each stage a doctrine at its most concrete and most collision-prone.
• 19 doctrines are tracked in assets/doctrine_risk_registry.json: 9 Critical, 8 High, 2 Medium, 0 Low — 17 of 19 doctrines (89%) require theologian review; the remaining 2 (Great Commission Fulfilled; Prayer and the Holy Spirit’s Guidance) require native speaker review. No doctrine in this curriculum is automated-review-only.
• The single highest-recurring collision risk is personal/relational divine categories being absorbed into impersonal or trance-based Chinese religious categories — the same underlying pattern the Romans baseline already documented for 天/天意/天命, now recurring across “filled with the Holy Spirit” (vs. mediumistic possession), “the Way” (vs. Daoist 道), providence language (vs. 天意/命运/缘分), and casting lots (vs. temple fortune-stick divination, 求签).

Risks

• 被圣灵充满 (filled with the Holy Spirit) and 这道 (the Way) are ranked the two highest-severity ambiguities in the entire curriculum (see 06_linguistic_gap_analysis.md §5): both sit one character-choice away from folk-religious spirit-possession and Daoist metaphysics, respectively, and both recur repeatedly (Pentecost narrative and Gentile Pentecost at Acts 10; five occurrences of “the Way”).
• 邪术 vs. 神迹奇事/神的大能 (condemned sorcery vs. God’s authenticating power): Chinese folk religion’s accepted category of ritual-practitioner power (法术) does not automatically register to readers as morally condemned, risking a neutral or even admiring reading of Simon Magus, Elymas, and the Ephesian magicians.
• 凡物公用 (held all things in common): uniquely for this curriculum, a biblical phrase collides with heavily loaded 20th-century Chinese political-economic vocabulary (collectivization), a risk with no parallel in the Romans baseline and no parallel in most other destination languages.
• 许愿 (vow) and 抽签 (casting lots): both reuse existing Chinese religious-practice vocabulary (temple vow-bargaining; fortune-stick divination) for biblical acts that must be taught as categorically different, requiring mandatory clarifying notes rather than relying on the term alone.
• Justification apart from the Law (Acts 13, 15) and the Resurrection/Ascension cluster (Acts 1–2, 17, 23) show the widest cross-tradition doctrinal divergence surveyed (04_comparative_theology.md), since nearly every non-Christian framework available to Mandarin readers (Confucian virtue-cultivation, Buddhist karma, folk merit-accounting, Daoist immortal-ascension) retains exactly the kind of earned-merit or cyclical-rebirth structure Acts repeatedly denies.

Opportunities

• Genuine, carefully-bounded bridge points exist and should be used pedagogically without claiming equivalence: Confucian loyalty-unto-death (忠臣死节) as a bridge for Persecution and Bold Witness; historical Buddhist missionary expansion into China as a bridge for the Great Commission’s cross-cultural transmission pattern.
• Acts’ own narrative structure does much of the doctrinal-distinguishing work already (e.g., the apostles explicitly refuse worship at Lystra and Paul does the same at Malta; Peter explicitly equates lying to the Spirit with lying to God) — Phase 2 translation and teaching notes need only preserve these corrective moves with full force rather than invent new argumentation.
• A single, well-enforced glossary (this Phase 1 output set) converts what would otherwise be 61 independent, error-prone judgment calls per translator into a fixed, auditable enforcement database, consistent with and extending the proven Romans baseline methodology.

Recommended actions

1. Load assets/translation_memory.json and assets/doctrine_risk_registry.json (Acts) alongside the baseline Romans equivalents before any Phase 2 segment translation begins; no exceptions.
2. Route every segment touching one of the 19 Critical-risk or 30 High-risk terms to mandatory human theologian review per the escalation rules in analysis/12_ai_translation_requirements.md; do not allow automated-only approval for any of these terms regardless of surrounding context.
3. Prioritize theologian review capacity for Acts chapters 2, 3, 5, 7, 8, 9, 10, 13, 14, 15, 16, 17, 19, and 28 — the highest Critical/High-risk-density chapters — ahead of predominantly narrative-historical chapters (18, 21, 25, 27) in Phase 2 scheduling.
4. Treat the 这道/道 fencing and the 被圣灵充满/spirit-possession fencing as a linked reviewer-training pair, since both concern the same underlying personal-vs-impersonal collision pattern already established in the Romans baseline.
5. Carry forward all 40 inherited baseline term entries without modification; any future baseline update to a shared term (e.g. 义, 圣灵, 恩典) must propagate automatically to this Acts package rather than being re-litigated here.